DeepSeek-R1-Distill-Qwen-1.5B is a large language model released by DeepSeek, released under the MIT license, with a claimed knowledge cutoff of 2025-01-20. Its weights, configuration, and tokenizer files have been cryptographically hashed and are queued for on-chain attestation by Weight Registry.
Description
DeepSeek's LLM series; R1 (2025) demonstrated strong reasoning performance at a fraction of prior assumed training cost, triggering significant real-world market reaction. This distilled variant compresses R1's reasoning into a small Qwen backbone.
Metadata
Hugging Face stats
470,297 downloads on Hugging Face · 1,591 likes · pipeline: text-generation · library: transformers
Weight Hashes
Weights file
SHA-25658858233513d76b8703e72eed6ce16807b523328188e13329257fb9594462945
SHA3-256f2591483b95759faff9044162fee6c2b69b40ac7f77876bc2f46aa625ed2a572
BLAKE38bc88a0c98c6472d12af929e0e303870ba678a2f32c89ca7500097227d514785
Config file
SHA-25637bd455e9679d2959536270fed49d25cc7c290a64f6e52abb97c71345a9cee41
SHA3-256e9fb71490f92ec0e2ba8a8f7e70b8c9bc6ca0ce1262a5358e4f178325d62276b
BLAKE3647c751925917c4d7f859b54102d5c3462d59e384e3a61a3bd0f1945226a1a48
Tokenizer file
SHA-25688145e3c3249adc2546ede277e9819d6e405e19072456e4b521cbc724bd60773
SHA3-2567c0feca25a5e57e7954519c86cca5ac89a64f278325973f910106387db096c11
BLAKE3723ab67cc74e50327da2cecc20b4c2891475d163b319d03af3709d7fc91732cb
Verify it yourself
curl -L -o "model.safetensors" "https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B/resolve/ad9f0ae0864d7fbcd1cd905e3c6c5b069cc8b562/model.safetensors" sha256sum "model.safetensors"
Expected SHA-256: 58858233513d76b8703e72eed6ce16807b523328188e13329257fb9594462945